Research into solar-assisted drying systems has been stepped up in response to the growing need for environmentally friendly drying technologies. These systems provide an alternative to traditional drying processes that is both cost-effective and beneficial to the environment. The utilization of solar energy, on the other hand, frequently faces challenges in the form of intermittent power and uncertain operating circumstances. The purpose of this research is to design a solar drying system that utilizes a dual working medium in order to increase the efficiency with which energy is utilized while simultaneously maintaining appropriate drying temperatures. Machine learning algorithms such as Bayesian Ridge, Linear Regression, Elastic Net, Support Vector Regression (SVR), and Gradient Boosting Regression Trees (GBRT) were utilized in order to forecast outlet air temperature depending on environmental parameters. This was done in order to improve performance prediction. According to the findings of the comparison, the GBRT model attained the highest accuracy, with an R² value of 0.98 during training and 0.94 during testing. This model outperformed other models, despite the fact that it required more time for training. The results of this study reveal that machine learning is a powerful tool that can accurately forecast the performance of solar drying systems, which in turn supports improved operational control and design optimization.